A Framework to Model ML Engineering Processes

TL;DR

This paper introduces a specialized framework and toolkit for modeling the complex, multidisciplinary processes involved in developing machine learning systems, aiming to improve communication and standardization.

Contribution

It presents a novel framework based on a domain-specific language tailored for ML engineering processes, addressing limitations of existing process modeling languages.

Findings

Framework facilitates better communication among ML teams

Toolkit supports practical implementation of the process models

Addresses gaps in current process modeling languages for ML development

Abstract

The development of Machine Learning (ML) based systems is complex and requires multidisciplinary teams with diverse skill sets. This may lead to communication issues or misapplication of best practices. Process models can alleviate these challenges by standardizing task orchestration, providing a common language to facilitate communication, and nurturing a collaborative environment. Unfortunately, current process modeling languages are not suitable for describing the development of such systems. In this paper, we introduce a framework for modeling ML-based software development processes, built around a domain-specific language and derived from an analysis of scientific and gray literature. A supporting toolkit is also available.